spatstat.core: plot.profilepl – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

plot.profilepl

Plot Profile Likelihood

Description

Plot the profile (pseudo) likelihood against the irregular parameters, for a model that was fitted by maximum profile (pseudo)likelihood.

Usage

## S3 method for class 'profilepl'
plot(x, ..., add = FALSE, main = NULL, tag = TRUE,
                         coeff = NULL, xvariable = NULL,
                         col = 1, lty = 1, lwd = 1,
                         col.opt = "green", lty.opt = 3, lwd.opt = 1)

Arguments

`x`	A point process model fitted by maximum profile (pseudo)likelihood. Object of class `"profilepl"`, obtained from `profilepl`.
`...`	Additional plot arguments passed to `plot.default` and `lines`.
`add`	Logical. If `TRUE`, the plot is drawn over the existing plot.
`main`	Optional. Main title for the plot. A character string or character vector.
`tag`	Logical value. If `TRUE` (the default), when the plot contains multiple curves corresponding to different values of a parameter, each curve will be labelled with the values of the irregular parameter.
`coeff`	Optional. If this is given, it should be a character string matching the name of one of the fitted model coefficients. This coefficient will then be plotted on the vertical axis.
`xvariable`	Optional. The name of the irregular parameter that should be plotted along the horizontal axis. The default is the first irregular parameter.
`col,lty,lwd`	Graphical parameters (colour, line type, line width) for the curves on the plot.
`col.opt, lty.opt, lwd.opt`	Graphical parameters for indicating the optimal parameter value.

Details

This is the plot method for the class "profilepl" of fitted point process models obtained by maximising the profile likelihood or profile pseudolikelihood.

The default behaviour is to plot the profile likelihood or profile pseudolikelihood on the vertical axis, against the value of the irregular parameter on the horizontal axis.

If there are several irregular parameters, then one of them is plotted on the horizontal axis, and the plot consists of many different curves, corresponding to different values of the other parameters. The parameter to be plotted on the horizontal axis is specified by the argument xvariable; the default is to use the parameter that was listed first in the original call to profilepl.

If coeff is given, it should be the name of one of the fitted model coefficients names(coef(as.ppm(x))). The fitted value of that coefficient is plotted on the vertical axis.

Value

Null.

Author(s)

Adrian Baddeley Adrian.Baddeley@curtin.edu.au, Rolf Turner r.turner@auckland.ac.nz and Ege Rubak rubak@math.aau.dk.

References

Baddeley, A. and Turner, R. (2000) Practical maximum pseudolikelihood for spatial point patterns. Australian and New Zealand Journal of Statistics 42, 283–322.

Baddeley, A., Rubak, E. and Turner, R. (2015) Spatial Point Patterns: Methodology and Applications with R. Chapman and Hall/CRC Press.

Examples

rstep <- if(interactive()) 0.005 else 0.025

  # one irregular parameter
  rr <- data.frame(r=seq(0.05,0.15, by=rstep))
  ps <- profilepl(rr, Strauss, cells)
  plot(ps)                      # profile pseudolikelihood 
  plot(ps, coeff="Interaction") # fitted interaction coefficient log(gamma)

  # two irregular parameters
  rs <- expand.grid(r=seq(0.05,0.15, by=rstep),sat=1:3)
  pg <- profilepl(rs, Geyer, cells)
  plot(pg) # profile pseudolikelihood against r for each value of 'sat'
  plot(pg, coeff="Interaction")
  plot(pg, xvariable="sat", col=ifelse(r < 0.1, "red", "green"))

spatstat.core

Core Functionality of the 'spatstat' Family

v2.1-2

GPL (>= 2)

Authors

Adrian Baddeley [aut, cre], Rolf Turner [aut], Ege Rubak [aut], Kasper Klitgaard Berthelsen [ctb], Achmad Choiruddin [ctb], Jean-Francois Coeurjolly [ctb], Ottmar Cronie [ctb], Tilman Davies [ctb], Julian Gilbey [ctb], Yongtao Guan [ctb], Ute Hahn [ctb], Kassel Hingee [ctb], Abdollah Jalilian [ctb], Marie-Colette van Lieshout [ctb], Greg McSwiggan [ctb], Tuomas Rajala [ctb], Suman Rakshit [ctb], Dominic Schuhmacher [ctb], Rasmus Plenge Waagepetersen [ctb], Hangsheng Wang [ctb]

Initial release

2021-04-17